AI Approaches to the Complexity of

Legal Systems AICOL International

Workshops 2015 2017 AICOL VI JURIX
2015 AICOL VII EKAW 2016 AICOL VIII
JURIX 2016 AICOL IX ICAIL 2017 and
AICOL X JURIX 2017 Revised Selected
Papers Ugo Pagallo
Visit to download the full and correct content document:
https://textbookfull.com/product/ai-approaches-to-the-complexity-of-legal-systems-aic
ol-international-workshops-2015-2017-aicol-vi-jurix-2015-aicol-vii-ekaw-2016-aicol-viii
-jurix-2016-aicol-ix-icail-2017-and-aicol-x-jurix-2017-revise/
More products digital (pdf, epub, mobi) instant
download maybe you interests ...

Real-Time Business Intelligence and Analytics:

International Workshops, BIRTE 2015, Kohala Coast, HI,
USA, August 31, 2015, BIRTE 2016, New Delhi, India,
September 5, 2016, BIRTE 2017, Munich, Germany, August
28, 2017, Revised Selected Papers Malu Castellanos
https://textbookfull.com/product/real-time-business-intelligence-
and-analytics-international-workshops-birte-2015-kohala-coast-hi-
usa-august-31-2015-birte-2016-new-delhi-india-
september-5-2016-birte-2017-munich-germany/

Business Information Systems Workshops: BIS 2017

International Workshops, Pozna■, Poland, June 28-30,
2017, Revised Papers 1st Edition Witold Abramowicz
(Eds.)
https://textbookfull.com/product/business-information-systems-
workshops-bis-2017-international-workshops-poznan-poland-
june-28-30-2017-revised-papers-1st-edition-witold-abramowicz-eds/

Euro Par 2017 Parallel Processing Workshops Euro Par

2017 International Workshops Santiago de Compostela
Spain August 28 29 2017 Revised Selected Papers 1st
Edition Dora B. Heras
https://textbookfull.com/product/euro-par-2017-parallel-
processing-workshops-euro-par-2017-international-workshops-
santiago-de-compostela-spain-august-28-29-2017-revised-selected-
papers-1st-edition-dora-b-heras/

Business Process Management Workshops: BPM 2017

International Workshops, Barcelona, Spain, September
10-11, 2017, Revised Papers 1st Edition Ernest Teniente

https://textbookfull.com/product/business-process-management-
workshops-bpm-2017-international-workshops-barcelona-spain-
september-10-11-2017-revised-papers-1st-edition-ernest-teniente/
English Legal System 2016 2017 17th Edition Catherine
Elliott

https://textbookfull.com/product/english-legal-
system-2016-2017-17th-edition-catherine-elliott/

Service Research and Innovation 5th and 6th

Australasian Symposium ASSRI 2015 and ASSRI 2017 Sydney
NSW Australia November 2 3 2015 and October 19 20 2017
Revised Selected Papers 1st Edition Amin Beheshti
https://textbookfull.com/product/service-research-and-
innovation-5th-and-6th-australasian-symposium-assri-2015-and-
assri-2017-sydney-nsw-australia-november-2-3-2015-and-
october-19-20-2017-revised-selected-papers-1st-edition-amin-behe/

Software Technologies Applications and Foundations STAF

2017 Collocated Workshops Marburg Germany July 17 21
2017 Revised Selected Papers 1st Edition Martina Seidl

https://textbookfull.com/product/software-technologies-
applications-and-foundations-staf-2017-collocated-workshops-
marburg-germany-july-17-21-2017-revised-selected-papers-1st-
edition-martina-seidl/

Multi Agent Systems and Agreement Technologies 15th

European Conference EUMAS 2017 and 5th International
Conference AT 2017 Evry France December 14 15 2017
Revised Selected Papers Francesco Belardinelli
https://textbookfull.com/product/multi-agent-systems-and-
agreement-technologies-15th-european-conference-
eumas-2017-and-5th-international-conference-at-2017-evry-france-
december-14-15-2017-revised-selected-papers-francesco-belardinel/

Computational Intelligence and Intelligent Systems 9th

International Symposium ISICA 2017 Guangzhou China
November 18 19 2017 Revised Selected Papers Part II
Kangshun Li
https://textbookfull.com/product/computational-intelligence-and-
intelligent-systems-9th-international-symposium-
isica-2017-guangzhou-china-november-18-19-2017-revised-selected-
 Ugo Pagallo · Monica Palmirani
Pompeu Casanovas · Giovanni Sartor
Serena Villata (Eds.)

AI Approaches
LNAI 10791

to the Complexity
of Legal Systems
AICOL International Workshops 2015–2017:
AICOL-VI@JURIX 2015, AICOL-VII@EKAW 2016,
AICOL-VIII@JURIX 2016, AICOL-IX@ICAIL 2017,
and AICOL-X@JURIX 2017, Revised Selected Papers

123
Lecture Notes in Artificial Intelligence 10791

Subseries of Lecture Notes in Computer Science

LNAI Series Editors

Randy Goebel
University of Alberta, Edmonton, Canada
Yuzuru Tanaka
Hokkaido University, Sapporo, Japan
Wolfgang Wahlster
DFKI and Saarland University, Saarbrücken, Germany

LNAI Founding Series Editor

Joerg Siekmann
DFKI and Saarland University, Saarbrücken, Germany
More information about this series at http://www.springer.com/series/1244
Ugo Pagallo Monica Palmirani
•

Pompeu Casanovas Giovanni Sartor

•

Serena Villata (Eds.)

AI Approaches
to the Complexity
of Legal Systems
AICOL International Workshops 2015–2017:
AICOL-VI@JURIX 2015, AICOL-VII@EKAW 2016,
AICOL-VIII@JURIX 2016, AICOL-IX@ICAIL 2017,
and AICOL-X@JURIX 2017
Revised Selected Papers

123
Editors
Ugo Pagallo Giovanni Sartor
University of Turin University of Bologna
Turin, Italy Bologna, Italy
Monica Palmirani Serena Villata
University of Bologna Inria - Sophia Antipolis-Méditerranée
Bologna, Italy Sophia Antipolis, France
Pompeu Casanovas
La Trobe University
Melbourne, VIC, Australia

ISSN 0302-9743 ISSN 1611-3349 (electronic)

Lecture Notes in Artificial Intelligence
ISBN 978-3-030-00177-3 ISBN 978-3-030-00178-0 (eBook)
https://doi.org/10.1007/978-3-030-00178-0

Library of Congress Control Number: 2018953366

LNCS Sublibrary: SL7 – Artificial Intelligence

© Springer Nature Switzerland AG 2018

This work is subject to copyright. All rights are reserved by the Publisher, whether the whole or part of the
material is concerned, specifically the rights of translation, reprinting, reuse of illustrations, recitation,
broadcasting, reproduction on microfilms or in any other physical way, and transmission or information
storage and retrieval, electronic adaptation, computer software, or by similar or dissimilar methodology now
known or hereafter developed.
The use of general descriptive names, registered names, trademarks, service marks, etc. in this publication
does not imply, even in the absence of a specific statement, that such names are exempt from the relevant
protective laws and regulations and therefore free for general use.
The publisher, the authors and the editors are safe to assume that the advice and information in this book are
believed to be true and accurate at the date of publication. Neither the publisher nor the authors or the editors
give a warranty, express or implied, with respect to the material contained herein or for any errors or
omissions that may have been made. The publisher remains neutral with regard to jurisdictional claims in
published maps and institutional affiliations.

This Springer imprint is published by the registered company Springer Nature Switzerland AG
The registered company address is: Gewerbestrasse 11, 6330 Cham, Switzerland
 Preface

AICOL stands for Artificial Intelligence Approaches to the Complexity of Legal

Systems. This volume presents the revised selected papers of the five different AICOL
Workshops during 2015–2017. The first took place as part of the JURIX 2015 con-
ference at the Universidade do Minho in Braga, Portugal, on December 9, 2015. The
second took place in Bologna, Italy, on November 19, 2016, in conjunction with the
20th International Conference on Knowledge Engineering and Knowledge Manage-
ment (EKAW 2016). The third was held at the JURIX 2016 conference at the Inria
Sophia Antipolis Mediterranée in Sophia Antipolis, France, on December 14, 2016, as
an AICOL Workshop. The fourth was the Workshop on MIning and REasoning with
Legal texts, held on June 16th, 2017, in London (UK) which was connected with the
MIREL (MIning and REasoning with Legal texts) project, H2020 Marie
Skłodowska-Curie grant agreement No 690974 (http://www.mirelproject.eu). It was
organized in conjunction with ICAIL 2017, the 16th International Conference on
Artificial Intelligence and Law. The fifth was held on December 13, 2017, at the
JURIX 2017 conference in Luxembourg. All these workshops had the common rational
to combine different disciplines for managing legal systems complexity, both in the
foundations and in the applications aspects. Two workshops in particular investigated
the Semantic Web techniques (EKAW 2016) and Natural Processing and Legal Rea-
soning models (MIREL 2016). The 37 selected papers, including the introduction,
represent a comprehensive picture of the state of the art in legal informatics.
The present volume follows the previous AICOL volumes: AICOL I-II, published in
2010, include papers from the first AICOL conference in Beijing (24th IVR Congress
in September 15–20, 2009, China) and the follow-up in Rotterdam (JURIX-09,
November 16–18, The Netherlands); AICOL III, published in 2012, resulting from the
third AICOL conference, held in Frankfurt a. M. (25th IVR, August 15–20, 2011,
Germany); AICOL-IV@IVR in Belo Horizonte Brazil, July 21–27, 2013, and
AICOL-V@SINTELNET-JURIX in Bologna, Italy, December 11, 2013.
Like its predecessors, this volume embodies the philosophy of the AICOL con-
ferences, which is to provide a meeting point for various researchers, such as legal
theorists, political scientists, linguists, logicians, and computational and cognitive
scientists, eager to discuss and share their findings and proposals. In this sense, the
keywords “complexity” and “complex systems” sum up the perspective chosen to
describe recent developments in AI and law, legal theory, argumentation, the Semantic
Web, and multi-agent systems.
AICOL incorporates in its VI edition the perspective of social intelligence, the
intertwined human–machine perspective on cognition, agency, and institutions. This
promising approach brings together the analytical and empirical perspectives of soci-
ety. Stemming from this starting point, the volume is divided into six main sections:
(i) Legal Philosophy, Conceptual Analysis, and Epistemic Approaches; (ii) Rules and
Norms Analysis and Representation; (iii) Rules and Norms Analysis and
VI Preface

Representation; (iv) Legal Ontologies and Semantic Annotation; (v) Legal Argumen-
tation; and (vi) Courts, Adjudication and Dispute Resolution. New entry topics need a
particular spotlight like Legal Design or Legal Data Analytics.
Finally, a special thanks is due to the excellent Program Committee for their hard
work in reviewing the submitted papers. Their criticism and very useful comments and
suggestions were instrumental in achieving a high quality of publication. We also thank
the authors for submitting good papers, responding to the reviewers’ comments, and
abiding by our production schedule.

July 2018 Ugo Pagallo

Monica Palmirani
Pompeu Casanovas
Giovanni Sartor
Serena Villata
 Organization

Organizing Committee
Danièle Bourcier Université de Paris II, France
Pompeu Casanovas Autonomous University of Barcelona, La Trobe
University
Monica Palmirani University of Bologna, Italy
Ugo Pagallo University of Turin, Italy
Giovanni Sartor European University Institute and University of Bologna,
Italy
Serena Villata Inria Sophia Antipolis, France

Program Committee
Laura Alonso Alemany Universidad Nacional de Córdoba
Michał Araszkiewicz Jagiellonian University
Guido Boella University of Torino
Daniele Bourcier Centre d’Etudes et de Recherches de Science
Administrative et Politique, Universite de Paris II
Pompeu Casanovas Autonomous University of Barcelona, La Trobe
University
Marcello Ceci GRCTC - Governance, Risk and Compliance Technology
Center
Pilar Dellunde Autonomous University of Barcelona
Luigi Di Caro University of Torino
Angelo Di Iorio University of Bologna
Enrico Francesconi ITTIG-CNR
Michael Genesereth Stanford University
Jorge Gonzalez-Conejero UAB Institute of Law and Technology
Guido Governatori Data61, CSIRO
Davide Grossi University of Groningen
John Hall Model Systems
Renato Iannella Queensland Health
Beishui Liao Zhejiang University
Arno R. Lodder Vrije Universiteit Amsterdam
Marco Manna University of Calabria
Martin Moguillansky Universidad Nacional del Sur
Paulo Novais University of Minho
Ugo Pagallo University of Torino
Marco Pagliani Senate of Italian Republic
Monica Palmirani University of Bologna
Adrian Paschke Freie Universität Berlin
VIII Organization

Silvio Peroni University of Bologna

Ginevra Peruginelli ITTIG-CNR
Enric Plaza IIIA-CSIC
Marta Poblet Autonomous University of Barcelona, RMIT University
Francesco Poggi University of Bologna
Martín Rezk Rakuten Inc.
Livio Robaldo University of Torino
Víctor Rodríguez Doncel Universidad Politécnica de Madrid
Antoni Roig Autonomous University of Barcelona
Piercarlo Rossi Università del Piemonte Orientale
Antonino Rotolo University of Bologna
Giovanni Sartor EUI, University of Bologna
Erich Schweighofer University of Vienna
Barry Smith SUNY Buffalo
Clara Smith Universidad Nacional de La Plata, Universidad Católica
de la Plata
Davide Sottara Arizona State University
Said Tabet EMC Corporation
Raimo Tuomela University of Helsinki
Leon van der Torre University of Luxembourg
Tom Van Engers University of Amsterdam
Anton Vedder Tilburg University
Serena Villata Inria Sophia Antipolis, Sophia Antipolis Cedex
Fabio Vitali University of Bologna
Radboud Winkels University of Amsterdam
Adam Wyner Swansea University
John Zeleznikow Victoria University
 Contents

Introduction: Legal and Ethical Dimensions of AI, NorMAS,

and the Web of Data . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
Ugo Pagallo, Monica Palmirani, Pompeu Casanovas,
Giovanni Sartor, and Serena Villata

Legal Philosophy, Conceptual Analysis, and Epistemic Approaches

RoboPrivacy and the Law as “Meta-Technology” . . . . . . . . . . . . . . . . . . . . 23

Ugo Pagallo

Revisiting Constitutive Rules . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 39

Giovanni Sileno, Alexander Boer, and Tom van Engers

The Truth in Law and Its Explication . . . . . . . . . . . . . . . . . . . . . . . . . . . . 56

Hajime Yoshino

From Words to Images Through Legal Visualization . . . . . . . . . . . . . . . . . . 72

Arianna Rossi and Monica Palmirani

Rules and Norms Analysis and Representation

A Petri Net-Based Notation for Normative Modeling: Evaluation

on Deontic Paradoxes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 89
Giovanni Sileno, Alexander Boer, and Tom van Engers

Legal Patterns for Different Constitutive Rules . . . . . . . . . . . . . . . . . . . . . . 105

Marcello Ceci, Tom Butler, Leona O’Brien, and Firas Al Khalil

An Architecture for Establishing Legal Semantic Workflows

in the Context of Integrated Law Enforcement . . . . . . . . . . . . . . . . . . . . . . 124
Markus Stumptner, Wolfgang Mayer, Georg Grossmann, Jixue Liu,
Wenhao Li, Pompeu Casanovas, Louis De Koker, Danuta Mendelson,
David Watts, and Bridget Bainbridge

Contributions to Modeling Patent Claims When Representing

Patent Knowledge . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 140
Simone R. N. Reis, Andre Reis, Jordi Carrabina,
and Pompeu Casanovas

Modeling, Execution and Analysis of Formalized Legal Norms

in Model Based Decision Structures . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 157
Bernhard Waltl, Thomas Reschenhofer, and Florian Matthes
X Contents

Causal Models of Legal Cases . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 172

Ruta Liepina, Giovanni Sartor, and Adam Wyner

Developing Rule-Based Expert System for People with Disabilities –

The Case of Succession Law . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 187
Michał Araszkiewicz and Maciej Kłodawski

Legal Vocabularies and Natural Language Processing

EuroVoc-Based Summarization of European Case Law . . . . . . . . . . . . . . . . 205

Florian Schmedding, Peter Klügl, David Baehrens, Christian Simon,
Kai Simon, and Katrin Tomanek

Towards Aligning legivoc Legal Vocabularies by Crowdsourcing . . . . . . . . . 220

Hughes-Jehan Vibert, Benoit Pin, and Pierre Jouvelot

Data Protection in Elderly Health Care Platforms . . . . . . . . . . . . . . . . . . . . 233

Angelo Costa, Aliaksandra Yelshyna, Teresa C. Moreira,
Francisco C. P. Andrade, Vicente Julian, and Paulo Novais

Assigning Creative Commons Licenses to Research Metadata:

Issues and Cases . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 245
Marta Poblet, Amir Aryani, Paolo Manghi, Kathryn Unsworth,
Jingbo Wang, Brigitte Hausstein, Sunje Dallmeier-Tiessen,
Claus-Peter Klas, Pompeu Casanovas, and Victor Rodriguez-Doncel

Dataset Alignment and Lexicalization to Support Multilingual Analysis

of Legal Documents . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 257
Armando Stellato, Manuel Fiorelli, Andrea Turbati, Tiziano Lorenzetti,
Peter Schmitz, Enrico Francesconi, Najeh Hajlaoui,
and Brahim Batouche

A Multilingual Access Module to Legal Texts . . . . . . . . . . . . . . . . . . . . . . 272

Kiril Simov, Petya Osenova, Iliana Simova, Hristo Konstantinov,
and Tenyo Tyankov

Combining Natural Language Processing Approaches for Rule Extraction

from Legal Documents . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 287
Mauro Dragoni, Serena Villata, Williams Rizzi, and Guido Governatori

Analysis of Legal References in an Emergency Legislative Setting . . . . . . . . 301

Monica Palmirani, Ilaria Bianchi, Luca Cervone,
and Francesco Draicchio

Legal Ontologies and Semantic Annotation

Using Legal Ontologies with Rules for Legal Textual Entailment . . . . . . . . . 317
Biralatei Fawei, Adam Wyner, Jeff Z. Pan, and Martin Kollingbaum
 Contents XI

KR4IPLaw Judgment Miner - Case-Law Mining for Legal

Norm Annotation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 325
Shashishekar Ramakrishna, Łukasz Górski, and Adrian Paschke

Towards Annotation of Legal Documents with Ontology Concepts . . . . . . . . 337

Kolawole John Adebayo, Luigi Di Caro, and Guido Boella

Reuse and Reengineering of Non-ontological Resources

in the Legal Domain . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 350
Cristiana Santos, Pompeu Casanovas, Víctor Rodríguez-Doncel,
and Leendert van der Torre

Ontology Modeling for Criminal Law . . . . . . . . . . . . . . . . . . . . . . . . . . . . 365

Chiseung Soh, Seungtak Lim, Kihyun Hong, and Young-Yik Rhim

ContrattiPubblici.org, a Semantic Knowledge Graph on Public

Procurement Information . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 380
Giuseppe Futia, Federico Morando, Alessio Melandri, Lorenzo Canova,
and Francesco Ruggiero

Application of Ontology Modularization for Building a Criminal

Domain Ontology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 394
Mirna El Ghosh, Hala Naja, Habib Abdulrab, and Mohamad Khalil

A Linked Data Terminology for Copyright Based on Ontolex-Lemon . . . . . . 410

Víctor Rodriguez-Doncel, Cristiana Santos, Pompeu Casanovas,
Asunción Gómez-Pérez, and Jorge Gracia

Legal Argumentation

Anything You Say May Be Used Against You in a Court of Law:

Abstract Agent Argumentation (Triple-A). . . . . . . . . . . . . . . . . . . . . . . . . . 427
Ryuta Arisaka, Ken Satoh, and Leendert van der Torre

A Machine Learning Approach to Argument Mining in Legal Documents . . . 443

Prakash Poudyal

Answering Complex Queries on Legal Networks: A Direct

and a Structured IR Approaches . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 451
Nada Mimouni, Adeline Nazarenko, and Sylvie Salotti

Inducing Predictive Models for Decision Support

in Administrative Adjudication . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 465
L. Karl Branting, Alexander Yeh, Brandy Weiss, Elizabeth Merkhofer,
and Bradford Brown

Arguments on the Interpretation of Sources of Law . . . . . . . . . . . . . . . . . . . 478

Robert van Doesburg and Tom van Engers
XII Contents

Courts, Adjudication and Dispute Resolution

Dynamics of the Judicial Process by Defeater Activation . . . . . . . . . . . . . . . 495

Martín O. Moguillansky and Guillermo R. Simari

Claim Detection in Judgments of the EU Court of Justice . . . . . . . . . . . . . . 513

Marco Lippi, Francesca Lagioia, Giuseppe Contissa,
Giovanni Sartor, and Paolo Torroni

A Non-intrusive Approach to Measuring Trust in Opponents

in a Negotiation Scenario . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 528
Marco Gomes, John Zeleznikow, and Paulo Novais

Network, Visualization, Analytics. A Tool Allowing Legal Scholars

to Experimentally Investigate EU Case Law . . . . . . . . . . . . . . . . . . . . . . . . 543
Nicola Lettieri, Sebastiano Faro, Delfina Malandrino,
Armando Faggiano, and Margherita Vestoso

Electronic Evidence Semantic Structure: Exchanging Evidence Across

Europe in a Coherent and Consistent Way . . . . . . . . . . . . . . . . . . . . . . . . . 556
Maria Angela Biasiotti, Sara Conti, and Fabrizio Turchi

Author Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 575

 Introduction: Legal and Ethical Dimensions
of AI, NorMAS, and the Web of Data

Ugo Pagallo1(&) , Monica Palmirani2 , Pompeu Casanovas3,4 ,

Giovanni Sartor2 , and Serena Villata5
1
Torino Law School, University of Torino, Lungo Dora Siena 100, 10153
Turin, Italy
ugo.pagallo@unito.it
2
CIRSFID, University of Bologna, via Zamboni 33, 40126 Bologna, Italy
{monica.palmirani,giovanni.sartor}@unibo.it
3
CRC D2D, La Trobe Law School, Bundoora Campus, Melbourne, Australia
pompeu.casanovas@uab.cat
4
IDT, Autonomous University of Barcelona, Bellaterra, Spain
5
INRIA Sophia Antipolis, Sophia Antipolis Cedex, France
serena.villata@inria.fr

Abstract. AICOL workshops aim to bridge the multiple ways of understanding

legal systems and legal reasoning in the field of AI and Law. Moreover, they pay
special attention to the complexity of both legal systems and legal studies, on
one hand, and the expanding power of the internet and engineering applications,
on the other. Along with a fruitful interaction and exchange of methodologies
and knowledge between some of the most relevant contributions to AI work on
contemporary legal systems, the goal is to integrate such a discussion with legal
theory, political philosophy, and empirical legal approaches. More particularly,
we focus on four subjects, namely, (i) language and complex systems in law;
(ii) ontologies and the representation of legal knowledge; (iii) argumentation and
logics; (iv) dialogue and legal multimedia.

Keywords: AI & law
Legal theory
Complex systems
Semantic web

Legal ontologies
Legal semantic web services
Argumentation

1 Introduction

The first volume of this workshop series on Artificial Intelligence approaches to the
complexity of legal systems (AICOL) was released at the very beginning of this decade
(2010). In the meanwhile, the field of Artificial Intelligence (“AI”) has known a new
renaissance: for instance, according to the tally Google provided to MIT Technology
Review in March 2017, the company published 218 journal or conference papers on
machine learning in 2016 alone, nearly twice as many as it did two years before [1].
Google’s AI explosion illustrates a more general trend that has to do with the
improvement of more sophisticated statistical and probabilistic methods, the increasing
availability of large amount of data and of cheap, enormous computational power, up to
the transformation of places and spaces into AI-friendly environments, e.g. smart cities

© Springer Nature Switzerland AG 2018

U. Pagallo et al. (Eds.): AICOL VI-X 2015–2017, LNAI 10791, pp. 1–20, 2018.
https://doi.org/10.1007/978-3-030-00178-0_1
2 U. Pagallo et al.

and domotics. All these factors have propelled a new ‘summer’ for AI. After the
military and business sectors, AI applications have entered into people’s lives. From
getting insurance to landing credit, from going to college to finding a job, even down to
the use of GPS for navigation and interaction with the voice recognition features on our
smartphones, AI is transforming and reshaping people’s daily interaction with others
and their environment. Whereas AI apps and systems often go hand-in-hand with the
breath-taking advancements in the field of robotics, the internet of things, and more, we
can grasp what is going on in this field in different ways [2]. Suffice it to mention in this
context two of them.
First, according to the Director of the Information Innovation Office (I2O) at the
Defense Advanced Research Projects Agency (DARPA) in the U.S. Department of
Defense, John Launchbury, there would have been so far two waves of research in AI.1
The first wave concerns systems based on “handcrafted knowledge,” such as programs
for logistics scheduling, programs that play chess, and in the legal domain, TurboTax.
Here, experts turn the complexity of the law into certain rules, and “the computer then
is able to work through these rules.” Although such AI systems excel at complex
reasoning, they were inadequate at perception and learning. In order to overcome these
limits, we thus had to wait for the second wave of AI, that is, systems based on
“statistical learning” and the “manifold hypothesis.” As shown by systems for voice
recognition and face recognition, the overall idea is that “natural data forms lower-
dimensional structures (manifolds) in the embedding space” and that the task for a
learning system is to separate these manifolds by “stretching” and “squashing” the
space.2
Along these lines, Richard and Daniel Susskind similarly propose to distinguish
between a first generation and a second generation of AI systems, namely, between
expert systems technologies and systems characterized by major progress in Big Data
and search [3]. What this new wave of AI entails has to do more with the impact of AI
on society and the law, than the law as a rich test bed and important application field for
logic-based AI research. Whether or not the first wave aimed to replicate knowledge
and reasoning processes that underpin human intelligence as a form of deductive logic,
the second wave of AI brings about “two possible futures for the professions,” that is,
either a more efficient version of the current state of affairs, or a profound transfor-
mation that will displace much of the work of traditional professionals. According to
this stance, we can thus expect that “in the short and medium terms, these two futures
will be realized in parallel.” Yet, the thesis of Richard and Daniel Susskind is that the
second future will prevail: in their phrasing, “we will find new and better ways to share
expertise in society, and our professions will steadily be dismantled” [3].
Against this highly problematical backdrop, three different levels of analysis should
be however differentiated. They concern the normative challenges of the second wave
of AI from a political, theoretical, and technical viewpoint, namely (i) the political
decisions that should be—or have already been—taken vis-à-vis current developments

1
See the video entitled “A DARPA Perspective on Artificial Intelligence,” available online at https://
youtube/-O01G3tSYpU.
2
Ibid.
 Introduction: Legal and Ethical Dimensions of AI, NorMAS, and the Web of Data 3

of e.g. self-driving cars, or autonomous lethal weapons; (ii) the profound transforma-
tions that affect today’s legal systems vis-à-vis the employment of e.g. machine
learning techniques; and, (iii) the advancements in the state-of-the-art that regard such
areas, as semantic web applications and language knowledge management in the legal
domain, or ejustice advanced applications. Each one of these levels of analysis is
deepening in the following sections.

2 Architectural Challenges

The political stance on current developments of AI hinges on a basic fact: the more the
second wave of AI advances, the more AI impacts on current pillars of society and the
law, so that political decisions will have to be taken as regards some AI applications,
such as lethal autonomous weapons, or self-driving cars. Over the past years, scholars,
non-profit organizations, and institutions alike have increasingly stressed the ethical
concerns and normative challenges brought about by many autonomous and intelligent
system designs [4–6]. The aim of the law to govern this field of technological inno-
vation suggests that we should distinguish between two different levels of political
intervention, that is, either through the primary rules of the law, or through its sec-
ondary rules [7].
According to the primary rules of the law, the goal is to directly govern social and
individual behaviour through the menace of legal sanctions. Legislators have so far
aimed to attain this end through methods of accident control that either cut back on the
scale of the activity via, e.g., strict liability rules, or intend to prevent such activities
through bans, or the precautionary principle. Regulations can be divided into four
different categories, that is, (a) the regulation of human producers and designers of AI
systems through law, e.g. either through ISO standards or liability norms for users of
AI; (b) the regulation of user behaviour through the design of AI, that is, by designing
AI systems in such a way that unlawful actions of humans are not allowed; (c) the
regulation of the legal effects of AI behaviour through the norms set up by lawmakers,
e.g. the effects of contracts and negotiations through AI applications; and, (d) the
regulation of AI behaviour through design, that is, by embedding normative constraints
into the design of the AI system [8].
Current default norms of legal responsibility can entail however a vicious circle,
since e.g. strict liability rules—let aside bans, or the precautionary principle—may end
up hindering research and development in this field. The recent wave of extremely
detailed regulations on the use of drones by the Italian Civil Aviation Authority, i.e.
“ENAC,” illustrates this deadlock. The paradox stressed in the field of web security
decades ago, could indeed be extended with a pinch of salt to the Italian regulation on
the use of drones as well: the only legal drone would be “one that is powered off, cast in
a block of concrete and sealed in a lead-lined room with armed guards – and even then
I have my doubts.” [9] As a result, we often lack enough data on the probability of
events, their consequences and costs, to determine the levels of risk and thus, the
amount of insurance premiums and further mechanisms, on which new forms of
accountability for the behaviour of such systems may hinge. How, then, can we prevent
legislations that may hinder the research in AI? How should we deal with the peculiar
4 U. Pagallo et al.

unpredictability and risky behaviour of some AI systems? How should we legally

regulate the future?
A feasible way out can be given by the secondary rules of the law, namely, the rules
of the law that create, modify, or suppress the primary rules of the system. Among the
multiple legal techniques with which we can properly address the normative challenges
of the second wave of AI, suffice it to mention here three of them. First, focus should be
on Justice Brandeis’s doctrine of experimental federalism, as espoused in New State Ice
Co. v Leibmann (285 US 262 (1932)). The idea is to flesh out the content of the rules
that shall govern individual behaviour through a beneficial competition among legal
systems. This is what occurs nowadays in the field of self-driving cars in the US, where
several states have enacted their own laws for this kind of technology. At its best
possible light, the same policy will be at work with the EU regulation in the field of
data protection [10, 11].
Second, attention should be drawn to the principle of implementation neutrality,
according to which regulations are by definition specific to that technology and yet do
not favour one or more of its possible implementations. The 2016 Federal Automated
Vehicles Policy of the U.S. Department of Transportation illustrates this legal tech-
nique. Although regulations are by definition specific to that technology, e.g. auton-
omous vehicles, there is no favouritism for one or more of its possible implementations.
Even when the law sets up a particular attribute of that technology, lawmakers can draft
the legal requirement in such a way that non-compliant implementations can be
modified to become compliant.
Third, legislators can adopt forms of legal experimentation. For example, over the
past decade and a half, the Japanese government has worked out a way to address the
normative challenges of robotics through the creation of special zones for their
empirical testing and development, namely, a form of living lab, or Tokku [12].
Likewise, in the field of autonomous vehicles, several EU countries have endorsed this
kind of approach: Sweden has sponsored the world’s first large-scale autonomous
driving pilot project, in which self-driving cars use public roads in everyday driving
conditions; Germany has allowed a number of tests with various levels of automation
on highways, e.g. Audi’s tests with an autonomously driving car on highway A9
between Ingolstadt and Nuremberg.
In general terms, these forms of experimentation through lawfully de-regulated
special zones represent the legal basis on which to collect empirical data and sufficient
knowledge to make rational decisions for a number of critical issues. We can improve
our understanding of how AI systems may react in various contexts and satisfy human
needs. We can better appreciate risks and threats brought on by possible losses of
control of AI systems, so as to keep them in check. We can further develop theoretical
frameworks that allow us to better appreciate the space of potential systems that avoid
undesirable behaviours. In addition, we can rationally address the legal aspects of this
experimentation, covering many potential issues raised by the next-generation AI
systems and managing such requirements, which often represent a formidable obstacle
for this kind of research, as public authorizations for security reasons, formal consent
for the processing and use of personal data, mechanisms of distributing risks through
insurance models and authentication systems, and more. The different legal techniques
and types of rules that lawmakers may employ, on the one hand, should not overlook
 Introduction: Legal and Ethical Dimensions of AI, NorMAS, and the Web of Data 5

the importance of the goals and values that are at stake with choices of technological
dependence, delegation and trust, in order to determine the good mix between legal
automation and public deliberation [13]. On the other hand, such choices of techno-
logical dependence, delegation and trust, through AI systems and procedures of legal
automation are affecting pillars and tenets of today’s law. As stressed above in this
introduction, AI technology profoundly affects both the requirements and functions of
the law, namely, what the law is supposed to be (requirements), and what it is called to
do (functions). This profound transformation has to be examined separately in the next
section.

3 Ethical and Legal Challenges: Device and Linked

Democracy

In one of the most celebrated 2014 John Klossner’s cartoons on the Internet of Things
the husband resignedly says to his wife: “We have to go out for dinner. The refrigerator
isn’t speaking to the stove.”3 This is not a joke anymore, and neither is the possibility
of connecting thousands of billions of devices that can literally speak to each other.
A world of smart objects shreds new challenges into the interconnected world of
humans and machines. The 2015 IBM Institute for Business Value Report [14] has
pointed out five major challenges: (i) the cost of connectivity (prohibitively high),
(ii) the Internet after trust (in the after-Snowden era “trust is over” and “IoT solutions
built as centralized systems with trusted partners is now something of a fantasy”),
(iii) not-future proof (many companies are quick to enter the market but it is very hard
to exit: the cost of software updates and fixes in products long obsolete and discon-
tinued), (iv) a lack of functional value (lack of meaningful value creation), (v) broken
business models (in information markets, the marginal cost of additional capacity—
advertising—or incremental supply—user data—is zero).
This is setting the conditions for “Device Democracy”, in which “devices are
empowered to autonomously execute digital contracts such as agreements, payments
and barters with peer devices by searching for their own software updates, verifying
trustworthiness with peers, and paying for and exchanging resources and services. This
allows them to function as self-maintaining, self-servicing devices” [14].
IBM suggests three new methodological trends for a scalable, secure, and efficient
IoT regarding: (i) architecture (private-by-design), (ii) business and economic insights
(key vectors of disruption), (iii) and product and user experience design (the trans-
formation of physical products into meaningful digital experiences).
The keyword here is the emergence of “meaningful experiences” in between
relationships, properties, and objects. Interestingly this has been also enhanced by 2016
and 2017 Gartner Hype Cycle of Emerging Technologies: (i) transparently immersive
experiences (such as human augmentation), (ii) perceptual smart machines (such as

3
https://www.computerworld.com/article/2858429/enterprise-applications/2014-the-tech-year-in-
cartoons.html#slide5.
6 U. Pagallo et al.

personal analytics), and (iii) digital platforms (including blockchain technologies).4

“AI everywhere” and blockchain are especially highlighted.
Our contention is that meaning is created, distributed, and framed through a
complex and dynamic world in which ethics and law cannot be set apart or let alone to
regulate processes, actions, and outcomes. Thus, regulations are entrenched and evolve
dynamically according to the network and the specific ecosystems they are contributing
to create. Meaningful experiences and smart devices entail smart regulations.
This is a feature already stressed for many years now by all attempts to frame new
developments in norMAS [15], non-standard deontic logic [16] and law and the
semantic web [17]. What is new is the attention brought to encompass innovation,
semantic developments and blockchain technologies with ethical, democratic (politi-
cal), and legal values alike.
In a recent Nexxus Whitepaper,5 Gavin Andresen, the leader of Blockchain
Foundation, equates crypto-currencies with empowering people: “an unstoppable
grass-roots movement that won’t be trampled on by any government or bank. It’s all
about the people’s freedom and reclaiming it” [18].
The first generation produced blockchain-based network protocols, such as Bitcoin,
Ethereum, Litecoin, and Zerocash. The second one can take this same idea of creating
encrypted blocks further: it uses distributed ledger databases along with user-
programmable smart contracts6 to create a number of social contracts in many fields—
universal basic income schemes, birth and death certificates, business licenses, property
titles, educational qualifications, marriage etc. This actually is the original dream of the
semantic web. MIT Digital Currency Initiative furnishes good examples in the public
domain, as they have launched the identity based services offered by BitNation, a
project aimed at decentralising governance at a global scale, e.g. a World Citizen-
ship ID based on blockchain, and a Refugee Emergency Response project. Ideally,
when applied to social or economic institutions, the result would be Decentralized
Autonomous Organizations (DAO): self-running organisations in which users can
become part of the chain performing things that computers cannot do [19].7
However, there is still a long road ahead before implementing DAO properly, as
complexity does not only lie on regulations, and transactions fuel a mixed, hybrid
social and economic reality that displays its own problems and conflicts. The recent
Ethereum crisis has shown security vulnerabilities [20]. Although preventive mecha-
nisms such as distributed consensus, cryptography, and anonymity are put in place,
iblockchain technologies remain vulnerable to many types of risks—mainly: the 51%
attack, account takeover, digital identity theft, money laundering, and hacking [21].

4
https://www.gartner.com/smarterwithgartner/.
5
Nexxus Partners was established in January 2016 in Texas, USA as a services company for the
bitcoin and cryptocurrency industry.
6
“Smart contracts are computer programs that can be correctly executed by a network of mutually
distrusting nodes, without the need of an external trusted authority” [20].
7
See the first Decentralized Autonomous Organization (DAO) code to automate organizational
governance and decision-making at [19].
 Introduction: Legal and Ethical Dimensions of AI, NorMAS, and the Web of Data 7

There are problems regarding the legal dimension as well. Some criticisms have
already stated that “cryptocurrencies cannot solve the problem of incomplete [rela-
tional] contracts, and as long as contracts are incomplete, humans will need to resolve
ambiguities” [22]. The same diffidence has been shown from a public legal standpoint,
as crypto-currencies cannot build up by their own a new public space [23].
The other way around, there are economic interpretations that highlight its positive
aspects. Ethereum and blockchain platforms have been received “as a new type of
economy: a ‘spontaneous organization’, which is a self-governing organization with the
coordination properties of a market (Hayek), the governance properties of a commons
(Ostrom), and the constitutional properties of a nation state (Brenan and Buchanan)” [24].
Perhaps this syncretic view is too over-confident, equating different dimensions
(economic, social, and legal) but, be as it may, law and the definition of law—what it
counts for—are at stake: “the legal status of DAOs remains the subject of active and
vigorous debate and discussion. Not everyone shares the same definition. […] Ulti-
mately, how a DAO functions and its legal status will depend on many factors,
including how DAO code is used, where it is used, and who uses it.” [19]
It is our contention that the synergy between different kinds of complementary
technologies can help to solve these regulatory puzzles and tensions, i.e. Blockchain is
the result of assembling two software paradigms (peer-to-peer applications and dis-
tributed hash tables). They are not the only ones. Semantic technologies and linked data
can be used to ease the tensions and create the shared scenarios in which crypto-
currencies and smart contracts can be safely and effectively used in a personalised
manner by a vast plurality of users. “Similarly as block chain technology can facilitate
distributed currency, trust and contracts application, Linked Data facilitated distributed
data management without central authorities” [25].
But for this to happen, to cross jurisdictions and different types of legal obstacles,
smart regulations and values are essential and should be similarly linked and har-
monised. Traditional legal tools at national, European and international levels, are
important, but they still fall short to cope with the complexity of algorithm governance
to reach metadata regulatory dimensions and layers [26]. This is why law, governance,
and ethics are at the same time being embedded into design, and re-enacted again as
contextually-driven to shape sustainable regulatory ecosystems. Beyond epistemic and
deliberative democracy, one of the concepts that have recently coined to describe this
new situation is linked democracy, i.e. the endorsement of (embedded) democratic
values to preserve rights and protect people on the web of data [27]. It is worth noticing
that these common trends are related to the combination of political crowdsourcing,
legal and ethical argumentation, and expert knowledge [28]. Innovation is deemed to
be a crucial component of democracy [29]. Thus, what the law is supposed to be and
what it is called to do are related not only to its architecture and tools (e.g. normative
systems, laws and rights) but to the many ways of balancing citizens’ compliance and
participation.
8 U. Pagallo et al.

4 Web of Data and Legal Analytics

Over the last two decades we have witnessed a remarkable volume of legal documents
and legal big data being put out in open format (e.g., the legal XML movement). The
information was represented using specific technical standards capable of modelling
legal knowledge, norms, and concepts [17, 30] in machine-readable format.
NormInRete [31] is an XML standard the Italian government issued in 2001 as the
official XML vocabulary for the country’s legislative documents. MetaLex was created
in 2002 in the Netherlands, and it evolved into CEN-MetaLex as a general format for
the interoperability of legal documents across Europe, this thanks to the EU Project
ESTRELLA [32]. Another significant outcome of the ESTRELLA project was the Legal
Knowledge Interchange Format-LKIF, composed of two main pillars: (i) a core legal
ontology [33] and (ii) a legal-rule language [34]. Even if these outcomes are encour-
aging, they lack a common-framework technical design making it possible to easily
integrate all the Semantic Web layers (e.g., text, norms, ontology). For this reason, the
Akoma Ntoso project [35] (an UNDESA-based African initiative)8 took the best
practices from those experiences and in 2006 designed a unique XSD schema for all
legal documents (e.g., including caselaw and UN resolutions [39]) and lawmaking
traditions (e.g., common law and civil law). In 2012, LegalDocML TC,9 of OASIS,
expanded the Akoma Ntoso XML vocabulary to embrace an international vision of
legal-document annotation. OASIS’s LegalCiteM10 TC provides semantic representa-
tion of legal references so as to foster a convergence of many existing syntaxes for legal
and legislative identifiers, including ELI [37], ECLI [36], URN-LEX,11 and the Akoma
Ntoso Naming Convention [40], making sure that legal document collections can
unambiguously be referred to and are also connectable to Linked Data assertions.
OASIS’s LegalRuleML TC [38] provides a standard for modelling constitutive and
prescriptive norms using formal language for rules. LegalDocML, LegalRuleML, and
LegalCiteM provide a common framework for modelling legal documents and for
fostering contextual metadata. The CLOUD4EU [41, 42] project offers a rare example
of a platform where those standards can act in an integrated manner: it is designed for
the General Data Protection Regulation (GDPR), making it possible to provide com-
pliance reports for this regulation.
LegalRuleML also provides an RDFS meta-model for modelling the deontic and
defeasible logic operators applied in the legal domain in order to export metadata in
RDF format. LegalDocML makes it possible to extract legal metadata and to convert it
into RDF. In the web of data paradigm, RDF triples produce a distributed and net-
worked legal knowledge repository that can be useful in enhancing the searchability of
relevant legal concepts, the semantic classification of documents, a light legal-
reasoning approach, and the integration of metadata with other nonlegal sources (e.g.,

8
United Nations Department of Economic and Social Affairs https://www.un.org/development/desa/
en/.
9
https://www.oasis-open.org/committees/legaldocml/.
10
https://www.oasis-open.org/committees/tc_home.php?wg_abbrev=legalcitem.
11
https://datatracker.ietf.org/doc/draft-spinosa-urn-lex/.
 Introduction: Legal and Ethical Dimensions of AI, NorMAS, and the Web of Data 9

DBpedia). However, the RDF technique is based on a collection of triples (assertions

composed of a subject, a predicate, and an object), often not validated, and the
underlying theory of inferential logic only approximates reality, like a map. In the
complex nonmonotonic legal system, RDF reasoning is quite risky, especially in situ-
ations where antecedents are rebutted (e.g., retroactivity, modifications of modifica-
tions, exceptions to exceptions, suspensions of the application of norms, annulment,
etc.). For this reason, the legal XML community is inclined to extract RDF assertions
as a subproduct of legal XML documents that are authentically, officially validated by
experts (e.g., principle of self-contained and self-explained assertions in a unique XML
file validated by the expert).
In the meantime, this large web of legal data connected with official digital legal
documents (often available from official-journal portals) in open format provides a rare
opportunity to apply artificial intelligence (AI) techniques in order to rethink the legal
theory and their institutions. In particular, legal analytics (LA)—combining data sci-
ence, artificial intelligence, machine learning (ML), natural language processing
techniques, and statistical methods—can reuse that vast and varied body of legal big
data, even if approximate, to infer new patterns and legal knowledge, so as to then
predict new models and bring out hidden correlations, often providing unexpected
insights into the relation between legal phenomena [43].
The automated application of learning methods to vast sets of examples makes it
possible to reproduce human behaviour and improve upon it, exploiting their com-
putational power to learn from successes and failures and thus improve performance.
Thanks to these new techniques and the sheer volume of the legal web of data now
available on the Internet, artificial intelligence has been able to advance from mock
examples to a host of real-life applications: conceptual retrieval and ranking, speech
and image recognition, question-answering, recommendations, translation, planning,
autonomous mobile robots, etc. Machine learning has been extensively applied to text
analytics, which refers to the use of multiple technologies—such as linguistic, statis-
tical, and machine learning techniques—to capture the information content of textual
sources. Considering that legal sources of all kinds are recorded in textual form (this is
true of statutes, regulations, judicial and other types of decisions and opinions, legal
doctrines, contracts), the application of text analytics to the law has huge potentials.
And indeed a number of applications are emerging, making it possible to automatically
classify and extract documents, identify principles in judicial decisions, and predict the
outcome of judicial cases.
Legal analytics (LA) has the potential to contribute to different aspects of legal
scholarship and practice. LA can advance legal informatics, making for a vast range of
successful new legal applications in the public and private sectors alike. In particular, it
supports the provision of AI applications, but overcoming the knowledge-acquisition
bottleneck. LA can support legal research, in particular, through its ability to unveil
hidden patterns in data and documents, revealing unseen features of the structure and
the functioning of legal institutions. The insights provided by LA, in combination with
computable models of the law, support the development of a new, empirically based
understanding of the law. LA can improve the efficiency and effectiveness of legal
institutions, while providing all legal actors with better knowledge of the law and of its
application. However, the use of big-data analytics has also raised a number of ethical
10 U. Pagallo et al.

issues that are already being discussed in several domains (targeted deceptive com-
mercial and noncommercial communication, discrimination, manipulation of public
opinion, etc.). Also already emerging in the law are some questionable practices,
particularly where law enforcement tries to predict illegal or otherwise unwanted
behaviour (e.g., a tendency to offend or reoffend). More generally, the knowledge
provided by analytics brings new ways to assess, influence, and control behaviour, and
these aspects have yet to be fully analysed. For instance, no study exists so far on how
the ability to predict court decisions, even those of specific judges, could influence
judicial decision-making.
The combination of LA and legal XML techniques improve our interpretation of
the AI inferences. XML nodes provide structural information and contextual metadata
that, in combination with the predictive assertions of LA, could be used to mitigate two
important negative side effects of LA techniques: (i) the introduction of bias from the
past experiences and mistakes (e.g., negative case-law, bad legal drafting practices in
legislation, influences due to socio-historical conditions) that can reinforce a tendency
to reiterate incorrect models and may impair the ability to creatively find brilliant new
solutions in the future (e.g., through filter-bubble effects); (ii) the fragmentation of legal
knowledge into separate sentences or isolated data without a logical connection making
it possible to achieve a consistent legal and logical narrative flow (e.g., contextless
prediction). XML nodes could provide the skeleton needed to reassemble the huge
amount of unexpected insights produced by the LA layer.
Finally, also crucial is the usability and the easy access to legal knowledge pro-
duced by LA. It is essential that the outcome of LA and legal XML sources in web
applications and new devices (e.g., augmented reality) be also understandable by
people who are not legal experts, without reframing the message; and, at any event, it is
also essential to provide clear mechanisms for explaining the algorithm decision-
making process and outcome. In this effort to achieve transparent communication we
can turn to human-computer interaction techniques, making it possible to create a fair
environment in which to better communicate the legal concepts and principles
extracted by LA. The legal design community is working to create new design patterns,
looking to provide better ways of displaying content, in such a way that the legal
community and end users (e.g., citizens) can place greater trust in LA and legal XML
[44–46].

5 On the Content of This Volume

This new AICOL volume is divided into six parts. They concern (i) legal philosophy,
conceptual analysis, and epistemic approaches; (ii) rules and norms analysis and rep-
resentation; (iii) legal vocabularies and natural language processing; (iv) legal
ontologies and semantic annotation; (v) legal argumentation; and, (vi) courts, adjudi-
cation and dispute resolution.
 Introduction: Legal and Ethical Dimensions of AI, NorMAS, and the Web of Data 11

5.1 Legal Philosophy, Conceptual Analysis, and Epistemic Approaches

In the first part of this volume, four papers deal with matters of legal philosophy,
conceptual analysis, and epistemic approaches. In RoboPrivacy and the Law as “Meta-
technology”, Ugo Pagallo examines how a particular class of robotic applications, i.e.
service robots, or consumer robots, may affect current legal frameworks of privacy and
data protection. Instead of a one-way movement of social evolution from technology to
law, a key component of the analysis concerns the aim of the law to govern techno-
logical innovation as well as human and artificial behaviour through the regulatory
tools of technology. By distinguishing between the primary rules of the law and its
secondary rules, e.g. forms of legal experimentation, the chapter illustrates some of the
ways in which the secondary rules of the law may allow us to understand what kind of
primary rules we may want for our robots. In Revisiting Constitutive Rules, Giovanni
Sileno, Alexander Boer and Tom Van Engers investigate how behaviour relates to
norms, i.e. how a certain conduct acquires meaning in institutional terms. By
addressing the double function of the ‘count-as’ relation, generally associated to
constitutive rules and mostly accounted for its classificatory functions, the chapter
reconsiders the relation between constitutive rules and regulative rules, and introduces
a preliminary account on the ontological status of constitution. In The Truth in Law and
Its Explication, Hajime Yoshino discusses what types of truth play their role in law and
in what way such types of truth can be explicated. Whereas the concept of truth in law
is classified into three types of truth, i.e. truth as fact, truth as validity and truth as
justice, the chapter provides their formal semantic foundation, and analyses both the
ways to explicate truth as validity and truth as justice in terms of intensional and
extensional explication, and how to grasp the reasoning of justification and of creation.
In From Words to Images Through Legal Visualization, Arianna Rossi and Monica
Palmirani discuss the process of sense-making and interpretation of visual legal con-
cepts that have been introduced in legal documents to make their meaning clearer and
more intelligible. Whilst visualizations have also been automatically generated from
semantically-enriched legal data, the analysis of current approaches to this subject
represents the starting point to propose an empirical methodology that is inspired by the
interaction with design practices and that will be tested in the future stages of the
research.

5.2 Rules and Norms Analysis and Representation

The second part of the volume, which has to do with rules and norms analysis and
representation, comprises seven contributions. In A Petri Net-based Notation for
Normative Modelling: Evaluation on Deontic Paradoxes, Giovanni Sileno, Alexander
Boer and Tom Van Engers focus on some of the problems that derive from the
development of systems operating in alignment with norms, e.g. the continuous flow of
events that modifies the normative directives under scrutiny. The chapter presents an
alternative approach to some of these problems, by extending the Petri net notation to
Logic Programming Petri Nets, so that the resulting visual formalism represents in an
integrated, yet distinct fashion, procedural and declarative aspects of the system. In
Legal Patterns for Different Constitutive Rules, Marcello Ceci, Tom Butler, Leona
12 U. Pagallo et al.

O’Brien and Firas Al Khalil illustrate a heuristic approach for the representation of
alethic statements as part of a methodology aimed at ensuring effective translation of
the regulatory text into a machine-readable language. The methodology includes an
intermediate language, accompanied by an XML persistence model, and introduces a
set of “legal concept patterns” to specifically represent the different constitutive
statements that can be found in e.g. financial regulations. In An Architecture for
Establishing Legal Semantic Workflows in the Context of Integrated Law Enforcement,
Markus Stumpner, Wolfgang Mayer, Pompeu Casanovas and Louis de Koker develop
a federated data platform that aims to enable the execution of integrated analytics on
data accessed from different external and internal sources, and to enable effective
support of an investigator or analyst working to evaluate evidence and manage
investigation structure. By preventing the shortcomings of traditional approaches, e.g.
high costs and silos-effects, the chapter also aims to show how this integration can be
compliant. In Contributions to Modelling Patent Claims when Representing Patent
Knowledge, Simone Reis, Andre Reis, Jordi Carrabina and Pompeu Casanovas
examine the modelling of patent claims in ontology based representation of patent
information. They relate to the internal structure of the claims and the use of the all-
element rule for patent coverage, in order to offer the general template for the structure
of the claim, and provide the visualization of the claims, the storage of claim infor-
mation in a web semantics framework, and the evaluation of claim coverage using
Description Logic. In Execution and Analysis of Formalized Legal Norms in Model
Based Decision Structures, Bernhard Waltl, Thomas Reschenhofer and Florian Matthes
describe a decision support system to represent the semantics of legal norms, whereas a
model based expression language (MxL) has been developed to coherently support the
formalization of logical and arithmetical operations. Such legal expert system is built
upon model based decision structures and three different components, namely a model
store, a model execution component, and an interaction component, have been worked
out, so as to finally test the execution and analysis of such structured legal norms vis-à-
vis the German child benefit regulations. In Causal Models of Legal Cases, Ruta
Liepina, Giovanni Sartor and Adam Wyner draw the attention to the requirements for
establishing and reasoning with causal links. In light of a semi-formal framework for
reasoning with causation that uses strict and defeasible rules for modelling factual
causation in legal cases, the chapter takes into account the complex relation between
formal, common sense, norm and policy based considerations of causation in legal
decision making with particular focus on their role in comparing alternative causal
explanations. In Developing Rule-Based Expert System for People with Disabilities,
Michał Araszkiewicz and Maciej Klodawski present the features of a moderately
simple legal expert system devoted to solving the most frequent legal problems of
disabled persons in Poland. By casting light on the structure of the expert system and
its methodology, the succession law of Poland and its procedures delivers sufficient
material to reveal the most important issues concerning such a project on a rule-based
expert system.
 Introduction: Legal and Ethical Dimensions of AI, NorMAS, and the Web of Data 13

5.3 Legal Vocabularies and Natural Language Processing

The third part of the volume regards legal vocabularies and natural language pro-
cessing. The eight contributions include EuroVoc-based Summarization of European
Case Law, in which Florian Schmedding illustrates the on-going development of a
multilingual pipeline for the summarization of European case law. By applying the
TextRank algorithm on concepts of the EuroVoc thesaurus, so as to extract summa-
rizing keywords and sentences, the intent is to demonstrate the feasibility and use-
fulness of the presented approach for five different languages and 18 document sources.
In Aligning Legivoc Legal Vocabularies by Crowdsourcing, Hughes-Jehan Vibert,
Benoit Pin and Pierre Jouvelot present the first Internet-based platform dedicated to the
diffusion, edition and alignment of legal vocabularies across countries. As a seamless
path for governments to disseminate their legal foundations and specify semantic
bridges between them, the chapter describes the general principles behind the legivoc
framework while providing some ideas about its implementation, e.g. crowdsourcing
the alignment of legal corpora together. In Data Protection in Elderly Health Care
Platforms, Ângelo Costa, Aliaksandra Yelshyna, Teresa C. Moreira, Francisco
Andrade, Vicente Julián and Paulo Novais deal with solutions to the increasing cog-
nitive problems that affect the elderly population in light of the iGenda project, which
aims to build safe environments that adapt themselves to one’s individual needs
through a Cognitive Assistant inserted in the Ambient Assisted Living area. Whereas
one of the main issues concerns the protection of the data flowing within the system
and the protection of the user’s fundamental rights, the chapter clarifies the principles
and legal guarantees of data protection, embracing appropriate solutions for techno-
logical features that may be a threat. In Assigning Creative Commons Licenses to
Research Metadata: Issues and Cases, Marta Poblet, Amir Aryani, Paolo Manghi,
Kathryn Unsworth, Jingbo Wang, Brigitte Hausstein, Sunje Dallmeier-Tiessen, Claus-
Peter Klas, Pompeu Casanovas and Víctor Rodriguez-Doncel tackle the problem of
lack of clear licensing and transparency of usage terms and conditions for research
metadata. Making research data connected, discoverable and reusable are the key
enablers of the new data revolution in research. Accordingly, the chapter does not only
discuss how the lack of transparency can hinder discovery of research data and make it
disconnected from publication and other trusted research outcomes. In addition, the
chapter suggests the application of Creative Commons licenses for research metadata,
and provides some examples of the applicability of this approach to internationally
known data infrastructures. In Dataset Alignment and Lexicalization to Support Mul-
tilingual Analysis of Legal Documents, Armando Stellato, Manuel Fiorelli, Andrea
Turbati, Tiziano Lorenzetti, Peter Schmitz, Enrico Francesconi, Najeh Hajlaoui and
Brahim Batouche tackle the complexity of the EU legal system, in which both the
linguistic and the conceptual aspects mutually interweave into a knowledge barrier that
is hard to break. In order to create a platform for multilingual cross-jurisdiction
accessibility to legal content in the EU, the chapter addresses the challenge of Semantic
Interoperability at both the conceptual and lexical level, by developing a coordinated
set of instruments for advanced lexicalization of RDF resources (be them ontologies,
thesauri and datasets in general) and for alignment of their content. In A Multilingual
Access Module to Legal Texts, Kiril Simov, Petya Osenova, Iliana Simova, Hristo
14 U. Pagallo et al.

Konstantinov and Tenyo Tyankov introduce a Multilingual Access Module, which

translates the user’s legislation query from its source language into the target language,
and retrieves the detected texts that match the query. More particularly, the unit con-
sists of two sub modules, i.e. an Ontology-based and a Statistical Machine Translation
units, which have their own drawbacks, so that both are used in an integrated archi-
tecture, in order to profiting from each other. In Combining Natural Language Pro-
cessing Approaches for Rule Extraction from Legal Documents, Mauro Dragoni,
Serena Villata, Williams Rizzi and Guido Governatori address the problem of moving
from a natural language legal text to the respective set of machine-readable conditions,
by combining the linguistic information provided by WordNet and a syntax-based
extraction of rules from legal texts, with a logic-based extraction of dependencies
between chunks of such texts. Such a combined approach leads to a powerful solution
towards the extraction of machine-readable rules from legal documents, which is
evaluated over the Australian “Telecommunications consumer protections code”. In
Analysis of Legal References in an Emergency Legislative Setting, Monica Palmirani,
Luca Cervone, Ilaria Bianchi and Francesco Draicchio provide for a taxonomy of legal
citations that set up an interesting apparatus for analysing a country’s legislative
approach. By investigating the references of a legal corpus of ordinances issued by the
Regional Commissioner for Emergency and Reconstruction over the first eighteen
months after the 2012 earthquake in Emilia-Romagna, the chapter scrutinizes the
critical issues arising in the regulative strategy for emergency situations. By distin-
guishing groupings based on lexical-textual analysis and groupings based on structural
element, the aim is to help lawmakers act better in future disasters, to extract infor-
mation concerning the number and the types of modifications produced, and to support
the debate on emergency national laws that deal with natural disasters.

5.4 Legal Ontologies and Semantic Annotation

As to the fourth part of this volume on legal ontologies and semantic annotation, it
comprises eight chapters. In Using Legal Ontologies with Rules for Legal Textual
Entailment, Biralatei James Fawei, Adam Wyner, Martin Kollingbaum and Jeff Z. Pan
describe an initial attempt to model and implement the automatic application of legal
knowledge using a rule-based approach. Whilst an NLP tool extracts information to
instantiate an ontology relative to concepts and relations, ontological elements are
associated with legal rules written in SWRL to draw inferences to an exam question.
Although further development of the methodology and identification of key issues
require future analysis, the preliminary results on a small sample are promising. In
KR4IPLaw Judgment Miner- Case-Law Mining for Legal Norm Annotation, Sha-
shishekar Ramakrishna, Łukasz Górski and Adrian Paschke offer a proof-of-concept
implementation for automatizing the process of identifying the most relevant judgments
pertaining to a legal field and further transforming them into a formal representation
format. On this basis, the annotated legal section and its related judgments can be
mapped into a decision model for further down the line processing. In Conceptual
Annotation of Legal Documents with Ontology Concepts, Kolawole John Adebayo,
Luigi Di Caro and Guido Boella illustrate a novel task of semantic labelling, which
exploits ontology in providing a fine-grained conceptual document segmentation and
 Introduction: Legal and Ethical Dimensions of AI, NorMAS, and the Web of Data 15

annotation. By dividing documents into semantically-coherent blocks and performing

conceptual tagging for efficient information filtering, the chapter presents a promising
solution, since the proposed task has several applications, such as granular information
filtering of legal texts, text summarization and information extraction among others. In
Reuse and Reengineering of Non-ontological Resources in the Legal Domain, Cristiana
Santos, Pompeu Casanovas, Víctor Rodríguez Doncel and Leon van der Torre explain
the processes of reusing and reengineering available non-ontological resources in the
legal domain into ontologies, taking into account their specificities, as they are highly
heterogeneous in their data model and contents. The description of these processes is
further clarified by using a case-study in the consumer law domain. In Ontology
Modelling for Criminal Law, Chiseung Soh, Seungtak Lim, Kihyun Hong and Young-
Yik Rhim propose a general methodology for designing criminal law ontologies and
rules. The chapter illustrates the super-domain ontology that contains the common
characteristics of criminal law, in order to explain the rule design method of criminal
law and present the application of the anti-graft act in Korea as an example. In Con-
trattiPubblici.org, a Semantic Knowledge Graph on Public Procurement Information,
Giuseppe Futia, Federico Morando, Alessio Melandri, Lorenzo Canova and Francesco
Ruggiero present the ContrattiPubblici.org project, which aims to develop a semantic
knowledge graph based on linked data principles in order to overcome the fragmen-
tation of existent datasets, to allow easy analysis, and to enable the reuse of infor-
mation. The objectives are to increase public awareness about public spending, to
improve transparency on the public procurement chain, and to help companies
retrieving useful knowledge for their business activities. In Application of Ontology
Modularization for Building a Criminal Domain Ontology, Mirna El Ghosh and Habib
Abdulrab carry out a survey on ontology modularization and present a modular
approach to build a criminal modular domain ontology (CriMOnto) for modelling the
legal norms of the Lebanese criminal system. CriMOnto, which will also be used for a
legal reasoning system, is composed of four independent modules that will be com-
bined together to compose the whole ontology. In A Linked Data Terminology for
Copyright Based on Ontolex-Lemon, Víctor Rodriguez-Doncel, Cristiana Santos,
Pompeu Casanovas, Asunción Gómez Pérez and Jorge Gracia discuss Ontolex-lemon,
namely, the de facto standard to represent lexica relative to ontologies that can be used
to encode term banks as RDF. A multi-lingual, multi-jurisdictional term bank of
copyright-related concepts has been published as linked data based on the ontolex-
lemon model. The terminology links information from several sources and the terms
have been hierarchically arranged, spanning multiple languages and targeting different
jurisdictions. Whilst the term bank has been published as a TBX dump file and is
publicly accessible as linked data, it has been used to annotate common licenses in the
RDFLicense dataset.

5.5 Legal Argumentation

Five chapters compose the fifth part of this volume on legal argumentation. In Abstract
Agent Argumentation (Triple-A), Ryuta Arisaka, Ken Satoh and Leon van der Torre
introduce a Dung style theory of abstract argumentation, which they call triple-A, in
which each agent decides autonomously whether to accept or reject her own arguments.
16 U. Pagallo et al.

By distinguishing between trusted arguments, selfish agents, and social agents, the
extensions of globally accepted arguments are defined using a game theoretic equi-
librium definition. In A Machine Learning Approach to Argument Mining in Legal
Documents, Prakash Poudyal analyzes and evaluates the natural language arguments
present in the European Court of Human Right (ECHR) Corpus. By dividing the
research into four modules, work on argumentative sentences vs. non-argumentative
sentences in narrative legal texts, is accomplished, so as to flesh out the features of this
module and conduct an experiment in Sequential Optimization Algorithm and Random
Forest Algorithm, which can be used as the basis of a general argument mining
framework. In Answering Complex Queries on Legal Networks: a Direct and a
Structured IR Approaches, Nada Mimouni, Adeline Nazarenko and Sylvie Salotti
compare two methods of search in legal collection networks, so as to present new
functionalities of search and browsing. Relying on a structured representation of the
collection graph, the first approach allows for approximate answers and knowledge
discovery, whilst the second one supports richer semantics and scalability but offers
fewer search functionalities. As a result, the chapter indicates how those approaches
could be combined to get the best of both. In Inducing Predictive Models for Decision
Support in Administrative Adjudicati, Karl Branting, Alexander Yeh and Brandy Weiss
explore the hypothesis that predictive models induced from previous administrative
decisions can improve subsequent decision-making processes. In light of three different
datasets, three different approaches for prediction in their domains were tested, showing
that each approach was capable of predicting outcomes. By exploring several
approaches that use predictive models to identify salient phrases in the predictive texts,
the chapter proposes a design for incorporating this information into a decision-support
tool. In Arguments on the Correct Interpretation of Sources of Law, Robert van
Doesburg and Tom van Engers deal with the formalization of legal reasoning and the
representation of law through computational models of argumentation. Whereas most
examples presented in literature can be characterized as post-hoc theory construction,
the chapter aims to provide an instrument that can be used to inform legal experts on
relevant issues in the process of solving current cases, i.e. using the interpretations of
legal sources ex-ante. An actual case that is in discussion in the Dutch Tax Admin-
istration, in court as well as in Parliament, helps to further clarify this approach.

5.6 Courts, Adjudication and Dispute Resolution

The sixth part of the volume is devoted to courts, adjudication and dispute resolution,
which are the subject matter of five chapters. In Dynamics of the Judicial Process by
Defeater Activation, Martin Moguillansky and Guillermo Simari illustrate a novel
activating approach to Argument Theory Change (ATC) for the study of the dynamics
of the judicial process. By considering the sentences of two different real criminal
procedures, the aim is to contribute to the discussion of how to deal with circumstances
of the judicial process like hypothetical reasoning for conducting investigations of a
legal case, and for handling the dynamics of the judicial process. In Claim Detection in
Judgments of the EU Court of Justice, Marco Lippi, Francesca Lagioia, Giuseppe
Contissa, Giovanni Sartor and Paolo Torroni address recent approaches to argumen-
tation mining in juridical documents, so as to present two distinct contributions. The
Another random document with
no related content on Scribd:
taken three and Freddie five. The caretaker was stoking up the fire at
the other side of the room.
“Ain’t you two goin’ ter join in the Musical Chairs?” he remarked,
contemplatively, “they’ve started ’em in the other room.”
Freddie took another ham-sandwich.
“I don’t feel extra like Musical Chairs,” he replied.
The caretaker grinned and shuffled out with the empty coke-
scuttle. It was precisely at that moment that Catherine began to
dislike the scent of Freddie’s lavishly spread hair-oil....
Catherine thought: “I don’t think I like him at all. I wonder if he
knows it was me who chalked up on the fence, ‘Freddie McKellar is a
soppy fool.’ ... ’Cos he is one, really....”
And then suddenly Freddie had an unfortunate inspiration.
He put his arm round her neck and touched her cheek. In an
instant she was up and flaring and standing before him.
“What on earth did you do that for?” she cried, passionately; “I
don’t want your smelly fingers on me!” (“Smelly fingers” was an
attribute she bestowed on everybody she disliked.)
He was astonished at her vehemence, but tried to carry it off
laughingly.
“Come back,” he called, advancing to her, “and don’t be silly ...
silly ... don’t be ... silly....” He was rather nervous. His nervousness
made him desperate.
There occurred a somewhat unseemly fracas. He stood before
the door and slowly got her trapped into a corner. She aimed a tea-
cup at him but missed. Maddened by this he rushed full tilt at her.
She struggled, snatched, tore, kicked, pinched. She was stronger
than he, but he got hold of her hair, and so held her at his mercy. He
just managed to kiss her. She spat in his face. Then he let her go.
She marched out of the room, seizing another tea-cup as she went.
When she was at the door she took a careful aim and flung it at him
with all her might. It struck his head. There was that tense pause just
after children are hurt, and just before they begin to cry. Then he
broke into a wail.... Most dramatically the piano in the next room
stopped, and there was the scuffle of finding chairs.... She paused at
the door and tossed her last words at him in uttermost scorn.
“Oh, you great big softie ...” she said, and passed out into the
cool night air.
She never enquired whether he were seriously hurt (he might
have been); she never stopped to think of the broken crockery on the
floor or her own red hair streaming in disarray; at that moment she
would not have cared if she had killed him.
And she never spoke to him again....
Afterwards she was doubly angry with him because he had made
her lose her temper....

§ 10
Mrs. Weston said: “Jus’ look at your hair! You’ve bin larking
abeaout, I darebebound.”
Catherine did not contradict her. “Larking about” was a
punishable misdemeanour.
“Everybody was larking about,” she put in, irrelevantly.
“You’re a disgrace,” continued Mrs. Weston, equally irrelevantly.
Then as an afterthought: “Larking about with the boys, I daresay....”
Catherine did not reply.
“Well, you’re going to have a sound thrashing, that’s all, so you
may as well know.... I’m about sick of your hooligan ways....”
Catherine went white. She was not afraid of a sound thrashing
(they were not very fearsome things when you got used to them); it
was the atmosphere of strained expectancy that was almost
intolerable. She went whiter when her mother said:
“Have your supper first.... There’s some cold rice pudding....”
 She ate in silence. Her mother was rushing in and out of the
scullery preparing her father’s supper. In the middle of all this her
father entered. He was tired and hoarse after the evening’s effort. He
noticed the strained atmosphere. He said to Catherine:
“What’s the matter, Cathie?”
Mrs. Weston began to talk very fast and very harshly. Her voice
was like the sudden rending of a strip of calico.
“She’s bin behaving herself badly again, that’s what’s amiss with
her.... Larking about all this evening, she was. A regular disgrace. I
tell you, I’m not going to put up with it. She’s going to get a sound
thrashing to teach her to remember....”
Simultaneously Mrs. Weston planked down a plate of greens and
vegetables in front of her husband. He attacked them nervously.
“It’s not good enough,” he said, after a pause, with the air of
being vaguely reproachful against nobody in particular, “I tell you it’s
not good enough.... I don’t know why these things should happen.
It’s not as if she was a little girl....”
That was all he said.
The sound thrashing began soon afterwards. It was an extremely
unscientific battery of slaps, in which Catherine dodged as best she
could amongst the crowded furniture of the kitchen. Once she
lurched against the table and knocked over the vinegar-bottle.
“I wish you wouldn’t ...” began Mr. Weston, and then stopped and
continued eating.
After some moments of this gymnastic display both parties were
hot and flushed with exertion, and the finale began when Mrs.
Weston opened the door of the lobby and manœuvred Catherine out
of the kitchen.
“Off you go,” she said. “Straight to bed ... str-h-aight to b-bed....”
The chase proceeded upstairs. Mrs. Weston’s stertorous
breathing and heavy footfalls were the most conspicuous sounds....
A few seconds afterwards a loud banging of an upstairs door
announced that hostilities were over.
In her tiny back bedroom Catherine sat down on a chair for
breath. She was not physically hurt; in her “larking about” with boys
and girls of her own age she had often paused for breath like this,
and at such times there had been joy in her heart even when there
had been pain in her body. But now she was conscious only of
profound indignity. Her father’s vague protest echoed in her memory:
“It’s not as if she were a little girl....”
She undressed and got into bed. It was quite dark, and she felt
acutely miserable. Far away the pumping engine at the water-works
whispered, as it always did at night-time, “Chug-chug ... chug-chug
... chug-chug-chug....” Ten, twelve years had passed since she had
counted the five knobs on the brass rail of the bed. She was growing
up, out of a child into a girl. She was not growing up without faults:
she knew that. The worst trait in her was temper ... she would have
to conquer that. She must learn self-control....
From below came the old familiar sound of her father taking off
his boots and dumping them under the sofa. “H-rooch-flop ...
h’rooch-flop.” ... That sound was bound up with all her memories of
childhood.
Ten minutes later there came a cautious tap at her door, and her
father entered in an intermediate stage of attire. He lit a candle
clumsily and shone it down upon her. She did not move. He prodded
her with his thumb in a vague, experimental way. She made no reply,
though her eyes were wide open and staring into his.
“I say, Cathie,” he began, vaguely and nervously, “you’ve bin
misbehaving, I’m told.... It’s too bad, you know.... Come now ... be a
good girl and go to sleep.”
Pause.
Then: “Kiss me.”
It was the first time for many years that he had asked for such a
thing. With no apparent reason at all the tears welled up into her
eyes, tears that she had hidden since her tenth birthday.
 She was just about to raise her head to meet his when a drop of
liquid candle grease fell on her bare arm. The sharp, unexpected
pain made her a prey to a sudden gust of tempestuous emotion....
“Oh, go away,” she muttered angrily, “don’t come bothering me ...
I’m tired....” She crouched down beneath the bedclothes with her
face turned away from him.
Mr. Weston retired a little sheepishly.
“Oh, well,” he said, “if you’re going to be sulky ... I suppose....”
When he had gone she cried as she had never cried before, and
all because she had spurned his proffered reconciliation. From the
other side of the thin partition that separated the two rooms, she
could hear the sharp “plock” as her father wrenched his collar off the
stud, and the steady nasal monotone of her mother’s voice. She
could not discern any words, but from the vicious way in which her
father kept stumbling up against things she guessed that they were
quarrelling....
 CHAPTER II
JEUNESSE
§1
CATHERINE won an open scholarship to the Upton Rising High
School for Girls. She did not win it because of any particular
brilliance or erudition in her examination papers; she won it, as a
matter of fact, because Mr. McGill, one of the Governors, happened
to remark to Miss Forsdyke, the headmistress: “I hear Weston’s got
his daughter in for a scholarship.”
Miss Forsdyke said, “Weston?—Weston?—Let me see—I believe
I’ve heard the name somewhere ... Er ... who is he?”
“One of the men at the Downsland Road School.... Not a bad
sort.... I bet old Clotters’ll be mad if Weston’s girl gets anything.
Clotters’ boy missed last year....”
Now Clotters was the headmaster of the Downsland Road
Council School. Mr. Weston did not like Mr. Clotters.... Mr. McGill did
not like Mr. Clotters.... And even Miss Forsdyke did not like Mr.
Clotters....
Thus it happened that Catherine obtained a scholarship to the
Upton Rising High School for Girls.
In her English paper she was asked to analyse: “There is a tide in
the affairs of men....” She began:
“There”—subject; “is” predicate; “a tide” object—according to a
well-established form of procedure which sometimes enabled her to
get her analysis right without in the least understanding what she
was about.
 And in her Scripture paper she was asked: What is a phylactery?
She answered: a kind of musical instrument.
Catherine was rather surprised to get a scholarship.

§2
Long lingering September evenings with the sun splashing over
the roofs of Upton Rising; the soft scented dusk creeping through
gravelled roads; tier upon tier of houses astride the hill, with every
window like a crimson star.... In the high road the newsboys were
calling, the trams swirled citywards like golden meteors flying
through space.... In the quiet residential roads was always the
chatter of the lawn-mower, the drowsy murmur of hedge-clipping....
In these delectable hours of twilight Catherine passed from Upton
Rising into Bockley.... Every night she passed, with swollen satchel
under her arm—Luke’s Grove, over Makepiece Common, then along
the Ridegway into the Bockley High Street.... And from the High
Street into Kitchener Road there was a bewildering choice of routes,
differing only in degrees of frowsiness....
Men passed her by like dim shadows heralded by the glowing
tips of their cigarettes....
The policeman on point-duty in the Bockley High Street knew her.
He said, “’d evenin’, miss,” and Catherine and the other girls who
accompanied her on her way home used to giggle hysterically, for he
was tall and handsome and presumably young.
Catherine went home with Madge Saunders and Helen Trant.
Madge was fat, good-natured, but lymphatic and uninteresting. Her
father was on the council and kept a big drapery stores in the High
Street. He called his daughter “Maggers,” and was excessively jovial
and contented. When Catherine went to tea at the Saunders’, he
called her “Carrots.” His humour was exhausted in the invention of
nicknames....
Helen Trant was almost the antithesis of Catherine and equally of
Madge. She was quiet, undemonstrative, but her quietness was not
the quietness of laziness. She worked hard, was moderately clever,
almost excessively conscientious, and in a quiet, unobtrusive way
immensely powerful and self-reliant. She was a scholarship girl, and
her father was in a good position in a London Insurance Office.
Neither Madge nor Helen was good-looking, but Helen had a quiet
dignity that made a fair substitute for beauty.... They were a rather
distinctive trio as they sauntered home together.
As they passed the policeman on point-duty Catherine made
provocative eyes at him. Madge rolled into heavy, undisciplined
laughter. And Helen sometimes smiled, but when she did it was the
smile as of one who knew all about policemen, their lives, wages,
conditions of existence, their baulked aspirations, confident hopes
and undying ambitions.... She looked to have the sympathy of one
who knows everything without being told anything....
Miss Forsdyke, in a spiteful mood, said:
“I wish, Helen, you would be more particular in your choice of
companions....”
Yet Catherine and Helen became close friends, and Madge was
merely an adjunct to their evening journeys home....

§3
Time was passing; Catherine was creeping through her teens,
and every night in the drawing-room at 24, Kitchener Road the piano
strummed for exactly one hour, and then stopped. By and by the
music-lover might have begun to detect certain tunes that were
familiar to him. A few of Mendelssohn’s “Songs Without Words,”
Tchaikovsky’s “Valse Triste,” the adagio part of Beethoven’s “Pathetic
Sonata.” ...
Once, too, a prelude of Chopin’s, chosen for its unChopin-like
qualities....
There came a day when Catherine’s playing began to be very
slightly superior to the instrument at her disposal. Nor did the latter
improve as time passed. All the lower notes responded with a nasal
twang reminiscent of a Jew’s-harp. The upper ones were so
physically inert that when pushed down they refused to come up
again without assistance, and so unanimous as to pitch that the
striking of the wrong note was no more inharmonious than the
striking of the right one....
Yet it was on this instrument that Catherine practised a certain
Fantasia in D Minor of Mozart’s that won her a first prize at the Upton
Rising Annual Eisteddfod.... The examiner was a wizened old man
with blue spectacles. From the first he annoyed Catherine. Her
music persisted in curling up.
“You should use a flat case,” he said, “not one of these roll ones.”
Then she discovered that the middle page of the music was not
there. Presumably she had left it in the waiting-room.
“You can’t go and fetch it,” he said. “I think you’re very
careless....”
“Do you?” she answered impudently. “Then I’ll play from
memory.”
“You ought not to play from memory ... at your age ...” he
protested.
Nevertheless she did so, and played better than she had ever
played at her practices. It was partly the ecstasy of manipulating a
splendid instrument, partly a reckless desire to defy and confute this
old man.
“H’m ...” he said, when she had finished. He bore her no malice
for her carelessness or impudence, he simply judged her fairly, totted
up her marks, and discovered them to be higher than the rest....
Accordingly he adjudged her the winner. He looked neither pleased
nor sorry.
Catherine decided that he was utterly soulless....

§4
 On a certain Monday morning Catherine and Helen took a day off
from school, and went picnicking in Epping Forest. Helen’s brother
George was with them, and also a friend of the latter’s, one Bert,
who took over the financial management of the outing with marked
efficiency, but was otherwise vague and indeterminate. George was
a moderately good-looking fellow of nineteen, clever in a restricted
kind of way, and very entertaining when the mood was upon him. He
worked for a City firm of accountants, and was taking his annual
fortnight’s holiday.
It was very pleasant to be strolling up the hill leading to High
Beech at ten o’clock on a fresh April morning. The party inevitably
split up into couples. Bert was walking on in front with Helen; George
and Catherine formed up the rear. There was a wonderful
atmosphere of serenity over everything: not a soul was about save
themselves; the hotels and refreshment châteaux seemed scarcely
to have wakened out of their winter’s sleep. And overhead the sky
was pure blue.... Up the steep, gritty road they trudged, and in the
hearts of each of them something seemed to be singing: “We are
going to have a glorious day.”
Bert was saying to Helen: “Yes, of course they’re very nice and
comfortable and all that, I know, but they fairly eat up the petrol....
Can’t possibly run them on less than ...”
“Indeed?” said Helen, sympathetically.
And a hundred yards behind them George was saying to
Catherine: “I suppose woman is a few inches nearer to mother earth
than man.... She is more ... primal ... no, not exactly that.... I mean
elemental ... that’s the word, I think....”
“She’s got more common sense, if that’s what you mean....”
“No, I don’t mean exactly that.... Besides, is common sense such
a virtue? ... The great things of the world have been done as a rule
by people with uncommon sense.... No, I mean this: woman seems
to know by instinct what man only learns by patient study and not
always then.... Isn’t that your experience?”
“I don’t think I’ve had any experience.”
 “H’m! ... the others are waiting for us at the top. I suppose they
want to know what we’re going to do....”
They quickened their steps to the summit.

§5
They chose for lunch a quiet spot hemmed in by ferns and
bushes. Catherine’s spirits soared higher and higher as the hours
flew.... The sun was splashing over the hills as they came upon the
red roofs of Chingford. The quantity of feeble, flippant conversation
that passed amongst them was colossal. But they had had a glorious
day....
“I’ll see you home,” said George, as they entered the straggling
outskirts of Bockley.
“Please don’t,” replied Catherine. “It’s quite out of your way.”
“I assure you ...” he began.
“Please ...” she reiterated. The truth was she did not wish her
mother to see her in the company of a young man.
Amidst the winedark fragrance of an April evening they passed
until they reached the corner of the road where the Trants lived.
They stopped talking here for three-quarters of an hour, and then
said good-bye. At the last minute George said:
“By the way, I’ve got to call in at a shop in the High Street to see
about something, so I may as well walk back part of the way with
you.”
Catherine blushed, but the darkness shielded her.
“The shops’ll be shut by this time,” said Helen, quietly.
“Er ... not ... er ... the shop I mean,” replied George.
He walked back with Catherine as far as the corner of High Street
and the Ridgeway. Their talk was rather vaguely, indefinitely
sentimental. Twice he quoted from Swinburne and once from Omar
Khayyám.
 As they descended the hill Catherine took off her tam-o’-shanter
hat and stuffed it in her pocket. The soft night breeze blew her hair
like a dim cloud behind her....
They shook hands in the dark interval between two brilliantly
lighted shop-windows.
“My God,” he whispered softly, “your hair!”
He brushed it lightly with his hand.
“What about it?” she said, and her voice was nearly as soft as
his.
“Passionate,” he cried; “like flame ... flame ... good-night....”
He fled into the dark vista of a side-street.

§6
The clock on the Carnegie library said, Ten minutes past ten.
Catherine thought, Now for a big row at home....
She had been forbidden to come in later than nine o’clock.
“When I was young ...” her mother had said.
And her father had argued: “I can’t see..what you need ever to be
out later than nine for.... You’ve got all the daytime, surely you don’t
need the night as well.... I can’t understand.... It’s not as if we didn’t
let you do what you like on Saturday afternoons....”
She put her hat on as she turned into Kitchener Road. She
sauntered slowly to No. 24. A minute or two won’t make much
difference, she reasoned, on top of an hour and a quarter. The
crowded memories of the day just past, coupled with anticipation of a
domestic fracas when she got home, combined to make her
somewhat excited. The day had been so full of incident that she
would have enjoyed walking the cool streets till midnight, reckoning
things out and sizing them according to their relative importance.
Then she recollected it was Monday night. Her father would be at
night-school; he did not usually arrive home till half-past ten.
 The street-lamp in front of No. 24 revealed the interesting fact
that the blinds in the front parlour were drawn. There was no light
behind them, but the tiny gas-jet in the hall was burning; she could
see its beam through the fanlight. Her heart leaped within her. She
felt like a prisoner granted a reprieve.... There were visitors. That
seemed certain. Somebody had come to spend the evening, and her
mother had “put a light in the front room,” the highest mark of respect
known. Now probably they were all having supper in the kitchen. The
hall-light, too, pointed to that conclusion, for ever since Mr. Tuppinger
took the wrong hat from the hall-stand, and failed to discover his
mistake afterwards, Mrs. Weston had made it a rule that the hall
should be illuminated when visitors came Catherine knocked at the
door.... This was really lucky. With good fortune the lateness of the
hour might not be noticed: at any rate the fracas would be
postponed. Also there would be a good supper awaiting her.... Cold
beetroot; perhaps even stewed prunes and custard....
A strange woman came to the door. Catherine did not know her
name, but she recognized her as someone who lived “up the road,”
and who used to push in front of her when she was a little girl at the
co-operative stores.
“Is it Cathie?” said the strange woman.
“Yes,” replied Catherine.
“Come inside,” answered the strange woman, with peculiar
solemnity. Then she went on, like the intoning of a chant:
“Your mother is not well ... in fact ... she’s had an accident ... in
the street ... in fact ... do come inside ... in fact....”
In fact, Mrs. Weston was dead.

§7
Mrs. Weston had been out shopping during the evening. In the
crowded part of the High Street she had been knocked down by a
bicycle. She had fallen upon her face, but had not apparently
received much hurt, for after having a cut attended to at the
chemist’s, she went home unattended. But at the very door of her
house in Kitchener Road, something went “snap” inside her head;
she collapsed and fell all in a heap on the doorstep. She was putting
the key in the lock when this happened, and the key was found in the
lock when neighbours came to her assistance. They carried her in
the front room (where the Collard and Collard piano was) and laid
her down on the sofa. She uttered vague scraps of conversation for
some moments: then she died....
When Catherine went in to look at her she could not help thinking
how death had made her look ridiculous. She was lying under the
window, and the lamp in the middle of the ceiling threw her features
into heavy shadow. There was a piece of sticking-plaster over the cut
on her forehead, and her chin was bruised as well. The most
prominent of her front teeth had broken off half-way, and as,
seemingly, she had died gasping for breath, her mouth was wide
open. The massive, almost masculine jaws hung unsymmetrically:
there was no beauty or calm in her last attitude. She looked as if she
had died fighting. An aperture in the drawn Venetian blinds allowed a
slit of pale light from the street lamp outside to cross her face
diagonally, making it appear more grotesque than ever. Catherine
could scarcely believe it was her mother. She had the old workaday
blouse on, because she had gone out shopping in a mackintosh and
had thought it would not show underneath. Catherine could not help
thinking how ashamed her mother would have been at the thought of
being seen in this blouse by all the neighbours, and especially to
have had the neighbours crowding in her own drawing-room with all
the cheap bamboo furniture and the faded carpet, and the “Present
from Margate” on the mantel-piece, and the certificate on the wall
certifying that John Weston, aged twelve, had achieved merit in
writing an essay on “Alcohol and its Effects on the Human Body.”
(This latter would have been removed long since, had it not
successfully covered up a hole in the wallpaper.) ... Catherine felt
sure that if her mother had known she was going to die, she would
have dressed up for the occasion. But it had come upon her
unexpectedly. There she was, with her shabby blouse and her
ghastly face, and her mackintosh and string-bag on the chair beside
her. There was some tea in the bag, and her fall had burst the paper
wrapper, for the latter was half-full, and there were tea grains about
the floor....
Mr. Weston had been sent for. He came in tired after a tiresome
day, plus the usual Monday feeling of discontent. He was in a bad
temper.
“Hell!” he muttered, as he bashed his shins against the piano in
the gloom. “These blinds ...” he began, and checked himself.
He seemed annoyed that she had done such a dramatic,
unexpected thing. He was annoyed that there was no supper ready
for him. “You might have got me a cup of tea ready,” he said to
Catherine. Then he tried to be conventional. “She was a good
woman,” he said, as if it had just occurred to him.
When the strange woman had departed, and Catherine and he
were sitting down in the kitchen to a frugal supper, he began the
conversation again.
“By the way,” he said, “apparently you didn’t go to school to-day.
Mrs. Jopson thought you’d be staying to the evening-class, and sent
a message to the school to fetch you. Miss Forsdyke said you hadn’t
been present at all to-day.... Is that so?”
“I didn’t go to-day,” admitted Catherine.
“Where did you go?”
“We ... took a day off ... picnicking in the Forest ... it seemed such
a fine day....”
“Who’s we?”
“Helen and ... and ... me.”
“Are you in the habit of taking days off like that?”
“Oh no.... It’s the first time we’ve ever done it.”
There was a pause.
“You know,” he went on protestingly, “this sort of thing’s not good
enough, Catherine.... You ought to see that this sort of thing can’t go
on ... it’s too bad of you ... running off to play truant ... and on the
very day that ... that your mother ...”
“How on earth could I——” she began hastily, and then stopped,
for she saw that big tears were rolling down both his cheeks.
“Not good enough,” he kept muttering, vaguely reproachful.
Then later on he reopened the question.
“I suppose—er—you and Helen were the only people at the
picnic?”
“No—there were two others.”
“Girls, I suppose?”
“No.”
“Not young men, I hope?”
“Yes, one of them was Helen’s brother. The other was a friend of
his....”
For a few moments he was very thoughtful. Then he continued:
“I don’t think you ought to have gone with them, Catherine ... at
your age, you know.... Besides, you’ve plenty of girl friends—I can’t
think what you want with young men and boys.... Girls should stick to
girls....”
“But surely, Father——”
“If you want friends, let them be girl friends ... surely you can find
plenty of your own sex without——”
Catherine could think of no adequate answer to this argument, so
she bade him good-night and went upstairs to bed....

§8
In the little back bedroom she sat down on the bed and tried to
gather her wits. She was overwhelmed by a feeling of physical
weariness: that was not surprising, for she had walked perhaps
fifteen miles that day. In the candle-light she saw her face in the
mirror: she was surprised to find herself almost ashen pale. Her red
hair floated cloud-like around her head: in the little hand-mirror there
was not room to see all of it at once. But it was still flying as if in the
wind, and it was gorgeously wild and untamed....
“My God,” George Trant had said, “your hair!” ...
... Catherine was surprised, almost shocked that she had as yet
shed no tears for her mother. It seemed such a brutally callous piece
of negligence, and Catherine was sure she was neither brutal nor
callous.... Yet tears would not come.
She undressed and got into bed....
The pumping-engine at the water-works went on at its patient
chug-chugging, and forthwith a myriad memories of childhood came
back to her.... She could feel the tears welling up into her eyes, and
then she realized that it was sentiment and not grief that was
affecting her. She would not weep for sentiment, like the heroines in
the six-penny novels that Madge Saunders read.
Ever and anon the whisper came echoing through her mind: “My
God ... your hair!” ...
From the very insistence of her thoughts she could not fall asleep
until morning was well advanced, but when she did, her sleep was
calm and dreamless....

§9
Of course there was a splendid funeral. It was infinitely more
gorgeous than anything that had taken place in Mrs. Weston’s
lifetime. Relatives were summoned to attend the obsequies, relatives
that Catherine had never seen and had not known existed, relatives
with black ties and rubicund faces and Cockney accents, and that
deplorable foreign flavour that comes of dwelling in another London
suburb. They all gathered together in the drab little front room
amongst the bamboo furniture, and gazed curiously at Catherine.
Evidently she did not quite realize their ideal of a bereaved daughter.
They were all a trifle nervous of the undertaker. Finally, they were all
squashed into four black coaches and driven slowly to the cemetery
behind a glass hearse. In front of the horses walked two men, each
bearing what appeared to be a mace.
The day was chilly and sprinkled with April showers; the
mourners in the first coach (in which was Catherine) insisted on
having all the windows closed, until the rain-washed panes were dim
with the reek of their breaths. They carried their pocket-
handkerchiefs in their hands, and spoke in tremulous murmurs....
The cavalcade swept on, through the dreariest and frowsiest streets
in all Bockley, out on to the murky highways where the mud splashes
from passing motor-buses reached the tops of the window-panes.
Then past the Town Hall, magnificently impeding traffic as they
crossed the tram-lines at the Ridgeway corner, on to the outer
fringes of the town, where public-houses and tin-missions indulged in
melancholy stares at one another across cat-haunted waste land. A
slow progress past an avenue of cars at the tram terminus, and at
last to the gates of a pretentious but infinitely dismal burial-ground.
The latter was owned and run on business lines by a limited liability
company, and for many years it had paid twelve per cent, on its
ordinary shares. That dying was a profitable industry could be seen
from the great gates, opening far back from the road, with their
ornate metal-work representing winged angels.
As they left the coaches a shower began. They walked about a
quarter of a mile amongst a welter of acrobatic angels, broken
columns and miscellaneous statuary; then they reached the grave.
The rain plashed dismally on the pile of brown earth by the side, and
everybody stood on the brink with a precarious footing on the
sodden soil. There was a diminutive Methodist parson with a bad
cold, who coughed at every comma in the burial-service and
sneezed into the grave at the end of each verse. All around them
was the litter of gravediggers’ tools, faded flowers and wreath-
skeletons. Catherine thought it by far the most depressing business
she had ever come across. Her father scattered a handful of cold,
clammy mud on top of the coffin, and everybody (especially the bald-
headed men with their hats off) seemed eager to get back to the fetid
warmth of the coaches.... So back went the procession, down the
long cemetery avenue, with nothing in sight save untidy vistas of
unsymmetrical gravestones, back into the steaming coaches, home
again through the mud and rain to Kitchener Road. The carriages
reeked with the smell of wet kid gloves and damp mackintoshes. In
the Bockley High Street they passed a crowd round a street
accident. A motor-bus had skidded into a tram-way standard, and
there were mud-splashed, white-painted ambulances in attendance.
Mr. Weston rubbed the vapour off the window with his hand. “Some
poor devil,” he muttered, and there was a whole world of humanity in
his voice. And Catherine felt that nothing in death itself was half so
terrible as the dismal fuss that people make over it.
When the carriages arrived back at No. 24, Kitchener Road, and
everybody went into the house, they found that the fire in the front
room had gone out. Half an hour was spent in trying to relight it with
damp coal and damp firewood and damp newspaper. Mr. Weston
held up the Bockley and District Advertiser to make a draught. The
newspaper caught alight and fell back on to the carpet, whereupon
Mr. Weston danced a sort of dervish cake-walk to stamp it out. This
acrobatic performance exercised a stimulating effect upon the
visitors, who became conversational. In a moment of riotous
abandon Mr. Weston directed Catherine to run over to the co-
operative stores and purchase two small tins of lobster and one large
tin of pineapple chunks....
About ten minutes to midnight, when all the mourners had
departed, and Catherine was pulling down the blinds in the back
bedroom, her father came up and sat down on the end of the bed
Unlacing his boots.
“You know, Cathie,” he began, nervously, as if there were
something he wished to get off his mind, “this business is so ... so ...
so sudden.... That’s what’s the matter with it. It don’t give a chap time
to gather his wits.... Last week she was here. Fussing about and
rushing round and seemingly in the best of health. And this week—
dead an’ buried.... Bit of a shock, isn’t it?”
She did not answer. He continued in a spurt:
 “You know there’s a sort of way in which you miss anybody
you’ve been used to seeing about the place for years an’ years.
Without any ... er ... what people call love, you know, or anything of
that sort.... Well, I miss your mother in that way. Quite apart from any
other way, I mean.... If she was here now she’d nag at me for not
taking my dirty boots off downstairs. It’s funny, but I shall miss all that
nagging. I got used to it. I didn’t particularly like it, but things’ll seem
pretty dull for a time without it....”
Pause.
“For twenty years I’ve chucked my dirty boots under the sofa
downstairs, and wouldn’t have dreamt of bringing them up here....
And now the first night she’s laid to rest I come up here with ’em on
without thinkin’ about it....”
He kept on making vague remarks.
“Life’s passing, Cathie ... one thing an’ then another.... Time waits
for no man—or woman.... We’re like those clocks at the railway
stations.... We seem not to be moving and then we fall forward with a
jerk at the end of the minute.... It’s easy to notice the jerks ... but time
goes steadily on whether we notice it or not....”
Then he changed the subject.
“It’s lucky for you it wasn’t an ordinary night last Monday, or you’d
have got in a fine row, I can tell you. Playing truant and going out
with young fellers.... A girl of your age ought not to bother her head
with fellers.... I never knew your mother till she was twenty-two....
This sort of free-and-easy-carrying-on won’t do, Catherine. For one
thing it’s not respectable. And for another thing it’s not right.... Find
some girl friends to go out with, and leave the fellers alone....”
“Fellers,” he called them. The word jarred on her.